Techniques for multipass welding
This disclosure provides systems, methods, and apparatuses, including computer programs encoded on computer storage media, that provide for welding techniques for manufacturing robots, such multipass welding techniques for welding robots. For example, the welding techniques may enable generation of weld instructions based on a welding fill plan. The instructions may be generated based on a bead model or a table that indicates a wire feed speed, a travel speed, or a voltage. As another example, the techniques may enable generation of weld instructions based on the one or more dimensions of a seam. As another example, the techniques may enable generation of a joint model of a cross-section of a seam to be welded. The joint model may be generated based on a combination of a plurality of feature components to generate the joint model of the seam. Other aspects and features are also claimed and described.
1 . A computer-implemented method of generating instructions for a welding robot, the computer-implemented method comprising:
receiving, by a robot controller of the welding robot, a computer aided design (CAD) model of a first part and a second part;
identifying, by the robot controller, a seam to be welded, the seam defined based on the first part and the second part of the CAD model;
discretizing, by the robot controller, a representation of the seam into a discrete waypoint set comprising a plurality of waypoints along a length of the seam, at least one waypoint of the plurality of waypoints constraining an orientation of a welding tool coupled to the welding robot in at least one degree of freedom;
for the plurality of waypoints along the length of the seam, determining, by the robot controller, a plurality of cross sections of a joint to weld the seam;
generating, with the robot controller, a welding fill plan for the seam, the welding fill plan generated based on at least the plurality of cross sections, a number of layers of the joint, a number of weld beads of the joint, and a bead model, wherein generating the welding fill plan includes:
determining a plurality of fill plan weld beads using machine learning logic, wherein for each weld bead of the plurality of fill plan weld beads, a target area and a torch angle of the welding tool is determined and the target area of the fill plan weld bead is adjusted between the plurality of cross sections along the length of the seam;
generating instructions, with the robot controller and based on the welding fill plan, for the welding robot to perform one or more weld passes;
transmitting, with the robot controller, the instructions to the welding robot; and
performing, by the welding robot, a welding operation including the one or more weld passes based on the instructions.
2 . The computer-implemented method of claim 1 ,
wherein for each weld bead of the plurality of fill plan weld beads, the torch angle of the welding tool is adjusted by the robot controller between the plurality of cross sections along the length of the seam.
3 . The computer-implemented method of claim 1 , further comprising:
generating, based on a plurality of feature components, a joint model of the plurality of cross sections; and
populating the joint model of the plurality of cross sections with a plurality of candidate weld beads, each candidate weld bead of the plurality of candidate weld beads having:
a size that is within a bead size range,
a position, or
a combination thereof,
wherein the plurality of fill plan weld beads is based on the plurality of candidate weld beads.
4 . The computer-implemented method of claim 3 , wherein populating the joint model with the plurality of candidate weld beads includes:
placing a first set of beads of the plurality of candidate weld beads, each bead of the first set of beads placed at a corresponding structural point at one or more structural points of the joint model.
5 . The computer-implemented method of claim 4 , wherein populating the joint model with the plurality of candidate weld beads includes placing:
after the first set of beads is placed, a second set of beads of the plurality of candidate weld beads at one or more auxiliary structural points of the joint model;
after the second set of beads is placed, a third set of beads of the plurality of candidate weld beads based on a cover profile of the joint model; and
after the third set of beads is placed, a fourth set of beads of the plurality of candidate weld beads at one or more unfilled spaces of the joint model.
6 . The computer-implemented method of claim 3 ,
wherein a number of beads of the plurality of candidate weld beads is different from a number of beads of the plurality of fill plan weld beads.
7 . The computer-implemented method of claim 1 , wherein the target area of each of the plurality of weld beads is adjusted between the plurality of cross sections along the length of the seam based upon dimension variability information of the seam.
8 . The computer-implemented method of claim 1 , further comprising:
receiving, with the robot controller, sensor data based on a weld bead physically deposited and formed by the one or more weld passes, the sensor data including data associated with formation of the weld bead, data associated with the formed weld bead, or a combination thereof.
9 . The computer-implemented method of claim 8 , further comprising:
updating the bead model using machine learning logic based upon the sensor data.
10 . The computer-implemented method of claim 1 , further comprising:
outputting, for display, welding fill plan information based on the welding fill plan, wherein the welding fill plan information indicates at least one of the number of layers, a number of beads in each layer, a bead size, a weld size, a cover profile, a cost of material, an average bead size, a minimum bead size, a maximum bead size, a distance of a bead from a structural point, or one or more cross-sections.
11 . A robotic welding system, comprising:
a welding robot; and
a robot controller comprising a processor and a non-transitory computer-readable medium storing code configured to be executed by the processor to cause the robot controller to:
receive a computer aided design (CAD) model of a first part and a second part;
identify a seam to be welded, including at least one dimensional variance along the seam, the seam defined based on the first part and the second part;
discretize a representation of the seam into a discrete waypoint set comprising a plurality of waypoints along a length of the seam, at least one waypoint of the plurality of waypoints constraining an orientation of a welding tool coupled to the welding robot in at least one degree of freedom;
determine a plurality of cross sectional weld profiles that differ based on the at least one dimensional variance;
generate a welding fill plan for the seam, the welding fill plan generated based on at least the plurality of cross sectional weld profiles and a bead model, wherein generating the welding fill plan includes:
determining a plurality of fill plan weld beads using machine learning logic, wherein for each weld bead of the plurality of fill plan weld beads, a target area and a torch angle of the welding tool is determined and the target area of the fill plan weld bead is adjusted between the plurality of cross sections along the length of the seam;
generate instructions, based on the welding fill plan, for the welding robot to perform a plurality of weld passes; and
perform, by the welding robot, a welding operation including the plurality of weld passes.
12 . The robotic welding system of claim 11 , wherein for each weld bead of the plurality of fill plan weld beads, the torch angle of the welding tool is adjusted by the robot controller between the plurality of cross sections along the length of the seam.
13 . The robotic welding system of claim 11 , wherein the plurality of weld passes deposit a weld having volumetric variation along the plurality of waypoints.